tariffs in nordic countries survey of load tariffs in … · tariffs in nordic countries – survey...
TRANSCRIPT
Tariffs in Nordic countries – survey of load tariffs in DSO grids
Report 3/2015
Tariffs in Nordic countries – survey of load tariffs in DSO grids
Nordic Energy Regulators
2015
2
Report 3/2015
NordREG c/o Danish Energy Regulatory Authority Valby 2500 Carl Jacobsens Vej 35 Denmark Telephone: + 45 41 71 54 00 Internet: www.nordicenergyregulators.org
November 2015
3
Table of contents
1 Introduction ................................................................................... 7
1.1 Background – Why a survey of tariff design? ................................. 9
1.1.1 Technological change – smart grid and meters ..................... 9
1.1.2 Energy Efficiency Directive .................................................... 9
1.2 Seminar with stakeholders ............................................................10
1.3 Outline of the report ......................................................................11
2 The Nordic design of load tariffs ............................................... 12
2.1 Implementation of the Energy Efficiency Directive in the Nordic
countries ................................................................................................12
2.1.1 Implementation of the directive in Denmark .........................12
2.1.2 Implementation of the directive in Finland ............................12
2.1.3 Implementation of the directive in Iceland ............................12
2.1.4 Implementation of the directive in Norway ............................12
2.1.5 Implementation of the directive in Sweden ...........................12
2.2 Tariffs in the Nordic countries .......................................................12
2.2.1 Tariffs for typical households ................................................13
3 Conclusions regarding differences and similaritiesFejl! Bogmærke
er ikke defineret.
3.1 The current structure ................. Fejl! Bogmærke er ikke defineret.
3.2 The trend for new design of tariffsFejl! Bogmærke er ikke defineret.
3.3 Policy options for regulators on tariff designFejl! Bogmærke er ikke
defineret.
4 Summary of studies done on load tariffs in DSO grids ........... 17
4.1 Load tariffs and demand response ................................................17
4.1.1 Norway: ................................................................................17
4.1.1 Sweden: ...............................................................................18
4.1.2 Finland: ................................................................................20
4.1.3 Denmark: .............................................................................21
4.2 What can be learnt about design of tariffs from the survey of studies
done? .....................................................................................................22
4
4.2.1 Cost-reflective tariffs ........................................................... 22
4.2.2 Demand response potential ................................................ 22
Annexes ............................................................................................. 23
5 Tariff design in Denmark ............................................................ 24
5.1 Few restrictions in how to design grid tariffs in Electricity Act ....... 24
5.2 Structure of grid tariffs for load in DSO-grids ................................ 25
5.3 Smart meters ............................................................................... 27
5.4 Discussions on tariff design in industry ........................................ 28
5.5 Conclusion ................................................................................... 30
6 Tariff design in Finland ............................................................... 31
6.1 Electricity Market Act ................................................................... 31
6.2 Structure of grid tariffs in DSO-grids............................................. 32
6.3 Discussions on tariff design ......................................................... 36
7 Tariff design in Iceland ............................................................... 38
7.1 Electricity Act ............................................................................... 38
7.2 Structure of grid tariffs in DSO- grids............................................ 38
7.3 Discussion on tariff design ........................................................... 39
7.4 Conclusion ................................................................................... 39
8 Tariff design in Norway .............................................................. 40
8.1 Electricity Act ............................................................................... 40
8.2 Structure of grid tariffs in DSO-grids............................................. 42
8.3 Discussions on tariff design ......................................................... 43
8.4 Conclusion ................................................................................... 46
9 Tariff design in Sweden .............................................................. 47
9.1 Limited restrictions on grid tariff design in the Electricity Act ........ 47
9.2 Structure of grid tariffs for load in DSO-grids ................................ 48
9.3 Discussions on tariff design in industry ........................................ 50
9.4 Conclusion ................................................................................... 52
5
Table over the report’s figures
Figure 1 | Fixed part of the grid tariff for a typical household customer in the Nordic
countries, 2015 .................................................................................................................. 14
Figure 2 | The price component for a typical household electricity customer in the Nordic
countries, 2015 .................................................................................................................. 15
Figure 3 | Fixed part of a grid tariff for different levels of consumption, 2015 ................ 26
Figure 4 | Elements of the electricity price in 2015 .......................................................... 27
Figure 5 | Implementation of smart meters in Denmark, 2013 ......................................... 28
Figure 6 | Daily load and limits for time differentiation, fictive example......................... 29
Figure 7 | Timedifferentiation of tariffs, fictive examples ................................................ 30
Figure 8 | Fixed part of the grid tariff for user type with 1×25 A, 2000 kWh/year in 2015
.......................................................................................................................................... 33
Figure 9 | Components of the retail electricity price, house with electric heating - 2.000
kWh/year ........................................................................................................................... 35
Figure 10 | Components of the retail electricity price, house with electric heating - 18.000
kWh/year ........................................................................................................................... 35
Figure 11 | Proportion of installed smart meters ............................................................... 36
Figure 12 | Components of electricity costs to a household, 2015 .................................... 38
Figure 13 | Fixed part of grid tariff for household customer with 20 MWh per year in
2015 .................................................................................................................................. 42
Figure 14 | Components of electricity costs to a household, 2015 .................................... 43
Figure 15 | Fixed part of the grid tariff for user types with 16 and 25 A, 5.000 and 20.000
kWh/year respectively in 2015 ......................................................................................... 49
Table over the report’s tables
Table 1 | Average typical household tariffs in the Nordic countries, 2015 ....................... 15
Table 2 | Smart meter implementation in the Nordic countries. ....................................... 16
Table 3 | Mean tariffs for different customer groups, 2015 .............................................. 26
Table 4 | Average tariffs by user type (in year 2015 price level, not including taxes) ..... 34
Table 5 | Proportion of fixed and variable parts of network tariff by user type ................ 34
Table 6 | Mean tariff for different customer groups in 2015 ............................................. 42
Table 7 | Proportion fixed charge in grid tariff for different customer groups ................. 43
Table 8 | Proportion fixed charge in grid tariff for different customer categories and
proportion of Time-of-use tariffs ...................................................................................... 49
6
Table 9 | DSO grids using power based tariffs for households in 2015. ........................... 50
7
1 Summary and conclusions regarding differences and similarities
1.1 The current structure
In the Nordic countries there is some variation in electricity network tariff design. In
Denmark, the fixed proportion of tariffs varies especially in the tariffs for the customer
group with low consumption. The fixed part of the tariff is usually bigger in Finland on
DSO’s operating in rural areas. The proportion of the fixed part of the tariff has been
increasing steadily in Finland. In Norway, the proportion of the fixed part of the tariff has
slightly increased from 2011 to 2015, but the fixed part of the tariff still constitutes a
smaller part of the total tariff. In Sweden, the variation in design is a typical feature of
tariffs. Still the cost structure is in principle the same, but there is a rather wide variation
in design in Sweden. The proportion of the fixed part of the tariff has over the years had a
slow increasing trend for a larger fixed part in Sweden too.
A development towards a larger proportion of fixed part of tariff can be seen in all Nordic
countries. There are though differences between the Nordic countries on the level of fixed
part’s proportion of the total tariff.
1.2 The trend for new design of tariffs
At the moment the design of tariffs has been left to the network companies to organise
within their income cap for all Nordic countires except Norway. Norway has some
regulations on allocating costs to fixed and variable parts of the network tariff, but even
here the DSOs have quite some degrees of freedom in setting the tariff. For the other
countires the network companies have their freedom to adjust tariffs according to their
individual cost structures and customer demographics. Pricing must still meet the specific
criteria of impartiality and non-discrimination, simplicity and transparency.
In Finland and Denmark, the industry has steered tariff design by guidance and tariffs are
quite unilateral throughout the network companies. The industry guidance contains
unambiguous guidance concerning all of these elements in the tariff methods. In general,
the current guidance is to include all DSO-specific cost types that depend on consumption
of electricity in the variable part of tariff and all cost types that do not vary with
consumption in the fixed part of the tariff. The guidance leaves DSOs to adjust tariffs
based on their individual cost structure. The trend on the other hand seems to be to
explore whether more capacity based tariff can be more cost reflectve instead of or in
addition to energy-based tariff.
1.3 Policy options for regulators on tariff design
In all the Nordic countries the role of national regulatory authority is to ensure that
pricing of network services is reasonable as a whole, so in other words within the revenue
8
cap. In Denmark, Finland, and Sweden there are no regulations on how the network
operators should allocate costs on the fixed and variable parts of network tariff. These
countries also share the following requirements concerning tariff setting:
- Tariffs must be cost-reflective,
- Tariffs must steer customers to efficient use of the power system and thus
encourage customers to save energy.
Legislation gives DSOs relatively free hands on tariff design but the regulator has to
ensure that these requirements are fulfilled in tariff setting. The role of the regulator is to
be neutral. When there is no legislation on the issue, in order for the regulatory authority
to take a position in favour on a certain type of tariff design, there should be clear benefits
compared to the previous design.
In recent years, industry-led working groups in the Nordic countries have been studying
the issue of tariff design. Norwegian national regulatory authority NVE has recently
formed a work group of its own. The work continues in 2016.
9
2 Introduction
2.1 Background – Why a survey of tariff design?
The premises for network regulation and tariff design are undergoing a rapid transition.
The implementation of the Energy Efficiency Directive in national legislation has lead to
increased requirements on tariffs, such as incentivising efficient grid usage. In addition to
new legislative conditions, the introduction of new technology opens up new
opportunities. Real time metering enables new types of tariffs, and monitoring and
steering equipment enables load steering and direct demand response. In parallel to these
changes, there is an increasing discussion about the need for harmonization of tariffs
between different countries.
Against this background, this report on tariff design produced by the NordREG Network
Regulation Working Group intends to answer the following question:
What can the NRA do to incentivize more energy efficiency in distribution and use of
energy through grid tariff design?
The report will provide a survey of the current DSO load tariffs in the Nordic countries,
present the ongoing discussions on tariff design, and assess the implementation of the
Energy Efficiency Directive in the countries as well as surveying the policy options for
NRAs to incentivise DSOs to design tariffs which are compatible with the Energy
Efficiency Directive.
2.1.1 Technological change – smart grid and meters
The development and introduction of new technologies provide new possibilities for tariff
design and network operation. Automatic meter reading (AMR) makes hourly metering
possible and enables a change from ordinary energy based charging to power based
charging as well as time-differentiated tariffs.
2.1.2 Energy Efficiency Directive
The Energy Efficiency Directive was implemented in 2012 and establishes a set of
binding measures to help the EU reach its 20 % energy efficiency target by 2020. Under
the Directive, all EU countries are required to use energy more efficiently at all stages to
the energy chain, from its production to its final consumption.
With regards to distribution operations and the topic of this report, the Energy Efficiency
Directive puts a number of new requirements on tariffs. These mainly relate to tariffs
which incentivise efficient use of the network. The Directive requires NRAs to provide
incentives for grid operators to make available system services to their users permitting
them to implement energy efficiency improvement measures. Specifically to tariff design,
the Directive requires that network tariffs are cost-reflective of cost-savings in the
network achieved from demand-side or demand-response measures and distributed
generation, including savings from lowering the cost of delivery or of network investment
and a more optimal operation of the network. In addition, the Directive specifies that
network tariffs may support dynamic pricing, for example through time-of-use tariffs or
critical peak pricing.
10
EU countries were required to transpose the Directive into national law by 5 June 2014. It
has been transposed in all Nordic EU member states.
2.2 Seminar with stakeholders
NordREG arranged a seminar in Stockholm on 5th November 2015 to discuss tariff design
with interested parties. The day consisted of presentations by 9 speakers and engaged
discussions with an active audience of around 50 people from all Nordic countries. The
speakers represented grid companies, industry associations, universities and consultant
agencies and were from Denmark, Finland, Norway and Sweden. They presented the
possibilities that can arise from new advanced meters and different alternatives for tariff
design. Some designs were theoretical and exploratory in their nature whereas others
were already implemented. Results from pilot projects with different tariff designs were
also presented.
New technology gives opportunities for providing customers with improved price signals
compared to traditional tariffs, with a fixed charge and a variable charge depending on
energy consumed. Among the participants in the seminar, there was a rather broad
consensus that price signals from tariffs are good in general. Numerous speakers
raisedthat tariffs should have a capacity element (as a supplement), as this would better
reflect the costs of the network. The network is dimensioned for peak capacity meaning
that capital costs are to some extent capacity-driven. Tariffs reflecting customers use
during the network’s maximum load are therefore relevant. Various participants
suggested making capacity-based tariffs time-dependent to better reflect the network
costs for the DSO.
Some challenges to providing price signals through tariffs to household customers were
raised. Designing a tariff model that sends price signals to customers that reflects when
capacity is scarce and that is also transparent for the end-user, and fairly easy to respond
to, is complicated. There may be a trade-off between sophisticated and cost-reflective
tariff models and transparency for the end user. However, improved tariff design was also
raised as a means to achieve better relations between the customers and the DSOs.
It was furthermore emphasized that the tariff is only one part of the end user’s electricity
cost. The electricity price is another important component and the price signals you want
to send through this may differ from the price signals of the tariffs. Some participants
agreed that the customer should only have to meet one price signal. Thus there may be a
reason to somehow coordinate price signals from the tariff and the electricity price.
Evidence was presented that suggested the customer needs a rather high saving potential
in order to change their behavior. There was also a discussion of the role of tariffs and to
what extent they should be designed with efficiency targets, climate goals and other
political issues in mind. The efficiency of tariffs as a tool for this was questioned.
Tariffs were discussed from the perspective of a supplier centric model by various
speakers. With suppliers being the main contact point for customers, the potential need
for harmonization was raised. The supplier centric model is however being implemented
differently in different countries, providing different starting points for DSOs when
designing their tariffs.
Many participants were pleased that NordREG is addressing the issue of tariff design, and
encouraged further work. Any changes in tariff design will redistribute costs between
customers; if some pay less, some must ultimately pay more. This must be communicated
11
to the customers, and it was seen as positive that NordREG and the NRAs would be part
of this communication. Participants also encouraged NRAs to explore what the customers
actually want. One conclusion from the seminar could be that the new technologies open
up possibilities, but it may not be useful to exploit all of them. Price signals that
consumers can respond to are good in principle, but their benefits must always be
weighed against their costs.
2.3 Outline of the report
The report describes the general design of load tarffis in the Nordic countries in chapter 3,
and chapter 4 summarizes studies done on load tariffs in the DSO grids in the different
Nordic countries. In the annexes, i.e. chapter 5-9, the Nordic countries describe in more
detail the different individual tariff setups in their respective countries, whithout drawing
similarities as such.
12
3 The Nordic design of load tariffs
3.1 Implementation of the Energy Efficiency Directive in the Nordic countries
3.1.1 Implementation of the directive in Denmark
The Energy Efficiency Directive is implemented in the Danish electricity act – most of
the directives content was already implemented in advance of the passing of the directive,
and the directive has not caused any changes in the electricity act regarding tariff setting.
3.1.2 Implementation of the directive in Finland
The energy efficiency directive has been implemented in Finland by the Energy
Efficiency Law. The Energy Efficiency Law entered into force 1 January 2015.
3.1.3 Implementation of the directive in Iceland
Iceland has not implemented the Energy Efficiency Directive since the Directive has not
yet been made part of the Agreement on the European Economic Area. At the moment
the government does not have any changes on the agenda as regards tariff design for the
Icelandic DSOs. The implementation of the Directive may, however, have some impact
on the legal framework that the DSOs currently work under in the future and indirectly
effect tariff design in Iceland.
3.1.4 Implementation of the directive in Norway
The Energy Efficiency Directive has not been implemented in the Norwegian electricity
act. There is an ongoing discussion about the Directive’s relevance for EEA.
3.1.5 Implementation of the directive in Sweden
The Energy Efficiency Directive has been implemented in the Swedish Electricity Act by
the parliament, among others through added provisions on efficient network operation.
This is described in further detail in the section “Tariff design in Sweden” in the Annex to
the report.
3.2 Tariffs in the Nordic countries
Overall the Nordic countries have different regulations on tariffs – cf. the annexes for
more specific information on the different countries. However, the principles on how to
regulate the tariffs are more or less the same. I.e. the pricing of electricity distribution
must meet specific defined criterias. Hereamong the tariffs must be impartial and non-
discriminatory for different customer groups, the tariffs must be simple and clear to
understand (i.e. transparent).
13
The network companies are solely responsible for designing tariffs within their income
cap (which is regulated by the regulatory authority) according to the regulation on tariff
structures. In the Nordic countries (except for Norway) there exists no regulation on how
the network operators should allocate costs on the fixed and variable parts of the network
tariff. In Norway there is regulation on how to allocate the fixed and variable costs; the
fixed component shall cover customer-specific costs and a share of the other fixed costs
of the network, while the energy component (variable cost) shall cover the cost of
marginal network losses and may in addition cover a share of the other costs not covered
by the fixed component.
For all countires the DSOs have an obligation to provide information about their tariffs to
the national regulatory authority.
In Denmark, Finland, Norway and Sweden the national regulatory authority does not
have a confirmation process regarding tariff structures of DSOs. The national regulatory
authority in Iceland has a confirmation process regarding tariff structures of DSOs.
In a case of disagreement on tariff structures between customer and DSO, National
Regulatory Authority in every country makes a decision if the tariffs are set according to
regulation.The regulatory approach is that price differentiation is in principle not allowed;
the Nordic countries, however, do have exemptions to this principle. For example price
differentiation for the sake of an effective use of the grid and security of supply is legal in
Denmark (the latter one from April 2016). In general there can be reasons to make
exemptions from this rule which would allow differentiated tariffs according to some
objective and reasonable criterias. In Norway, tariffs can be differentiated according to
objective and verifiable criteria based on relevant grid conditions.
For example there exists tariff differentiation on different customer groups in all the
Nordic countries. The specific customer groups in the different Nordic countries vary,
however, the general customer grouping can be grouped into the following three customer
categories household, small industry and large industry customers.
3.2.1 Tariffs for typical households
The tariffs for household customers in each country vary in design even though the
structure of the cost in principle is the same.
The following consumption types represent a typical household in the Nordic countries:
Denmark: Annual consumption 4.000 kWh
Finland: Annual consumption 5.000 kWh and 20.000 kWh
Norway: Annual consumption 20.000 kWh
Sweden: Annual consumption 5.000 kWh and 20.000 kWh
Several observations can be derived from Figure 1. First of all, Sweden has the highest
number of DSOs. On average the fixed share of network tariff is higher in Sweden than it
is in other countries. This is true with consumption types with lower annual consumption
(4.000 – 5.000 kWh) and higher annual consumption (20.000 kWh). DSOs in Norway on
average have lowest fixed part of the grid tariff. Denmark and Finland more or less
follow the same pattern regarding fixed part’s share of the total grid tariff, when the
number of DSOs is increased although the share of fixed part is on average higher in
14
Denmark. Regarding fixed part’s share of the network tariff on average, there is not so
much difference in Finland between consumption types than there is in Sweden.
Average fixed and variable shares of network tariff for household customers in Denmark,
Finland, Norway and Sweden are listed in Table 1.
Figure 1 | Fixed part of the grid tariff for a typical household customer in the Nordic countries, 2015
Source: DERA, Energiavirasto, NVE & Swedish Energi Market Inspectorate
The amount which tariffs account for in the total electricity bill varies between the Nordic
countries, cf. Figure 2. However, in general the tariffs account for maximum of one third
of the electricity bill with the exception of Iceland where network tariff account for
approximately half of the electricity bill. The cost of electricity power is more or less one
third of the total electricity price with the exception of Denmark, where it is 18 percent.
In Finland the electricity power’s share of the total electricity bill is highest in the Nordic
countries. Fx. in Denmark the network tariff only account for around 15 percent of the
electricity bill, but this is due to high electricity taxes in Denmark (67 percent). The share
of electricity taxes is lowest in Iceland.
15
Figure 2 | The price component for a typical household electricity customer in the Nordic countries, 2015
Source: DERA, Energiavirasto, Orkustofnun, Statistisk Sentralbyrå & Sveriges el- och naturgasmarknader 2014
The average network tariff for a typical household in the Nordic countries is shown in
Table 1. The average annual network tariff is highest in Sweden. Also, the fixed part of
the network tariff on average is highest in Sweden. This is the case for both consumption
types. Denmark, Finland and Norway are closer to each other regarding the average
annual cost of network tariff when differences in consumption types are taken into
account. Regarding the variable part of tariff, the cost per kWh seems to decrese when
annual consumption increases. The cost per kWh in the variable proportion of tariff is
lowest in Denmark when we look on the consumption type with low annual consumption.
When we look on the consumption type with high annual consumption, the cost per kWh
in the variable component of the tariff is lowest in Norway.
Table 1 | Average typical household tariffs in the Nordic countries, 2015
Denmark € (DKK)
Finland, € Norway € (NOK)
Sweden, € (SEK)
Consumption type (kWh/year)
4.000 5.000 20.000 20.000 5.000 20.000
Average annual network tariff
178 € (1.327)
224 € 505 € 583 €
(5.247) 314 €
(2.924) 680 €
(6.326)
Variable component of tariff on average, €/kWh
0,027 € (0,2)
0,045 € 0,025 € 0,020 €
(0,18) 0,063 €
(0,58) 0,034 €
(0,32)
Fixed part of annual network tariff on average €
70 € (525)
99 € 188 € 181 €
(1.627) 228 €
(2.122) 348 €
(3.242)
Fixed part on average % 40 % 44 % 37 % 31 % 73 % 51 %
Variable part on average % 60 % 56 % 63 % 69 % 27 % 49 %
Source: DERA, Energiavirasto, NVE & Swedish Energi Market Inspectorate
Smart meter implementation is more or less fully implemented in Finland and Sweden,
while Denmark has implemented smart meters for roughly half of its electricity
customers. Norway has 10 percent while Iceland has not yet implemented any smart
meters. So the diversity of smart meter implementation is high across the Nordic
countries, cf. Table 2.
16
Table 2 | Smart meter implementation in the Nordic countries.
Denmark Finland Iceland Norway Sweden
57 % 95 % 0 % 10 % 100 %
Source: Danish Energy Association, Energiavirasto, NVE & Swedish Energi Market Inspectorate Note: Time of date for the smart meter implementation was in respective order 2013, 2014, 2015, 2014 and 2009.
17
4 Summary of studies done on load tariffs in DSO grids
The ongoing developments in tariff design and network regulation are receiving
substantial attention both from commercial parties and academia. To provide a
background to the survey and analysis of current tariff design in the Nordic countries, this
chapter presents a survey of research and previous studies performed on the topic in the
respective countries.
4.1 Load tariffs and demand response
4.1.1 Norway:
“Optimal network tariffs and allocation of cost” (Report 2008-129) by Econ.
http://www.nve.no/Global/Kraftmarked/Tariffer/Optimal%20network%20tariffs%20and
%20allocation%20of%20costs.pdf?epslanguage=en
“Pricing of transmission and distribution of electricity should be done according to
economic criteria, that is reflect the marginal short-term costs of losses and congestion in
the grid. Long-term price signals beyond these short-term signals should reflect the cost
of customer-specific investments. As the grid constitutes a natural monopoly, these tariffs
will however not be sufficient to cover the total need for income in the grid. Hence, there
will be a need for tariffs that provide recovery of residual costs. The allocation of
residual costs between network customers should be done in a manner which distorts use
of the grid and investments as little as possible. In practice, this means relatively low
tariffs for generation and large industrial users, while households, the public sector and
small businesses should cover the bulk of the residual costs, although there are many
variations within these groups and over time. The Norwegian tariff system incorporates
many of the economically correct principles, but it is practically impossible to implement
a theoretically perfect system.”
Future design of loadtariffs (“Framtidig utforming av nett-tariffer”) (Report from EC-
Group, 2014).
“The report aims for a cost-reflective design for load tariffs. Based on the cost structure
for the Norwegian network, the allocation of income between the different tariff
components should be: Approximately 20% from energy component, approximately 10%
from the part of the fixed component that covers customer-specific costs, approximately
20% from the part of the fixed component that covers costs mainly in the low-voltage
network and approximately 50% from the part of the fixed component that covers
administrative costs and costs in the overall parts of the network (mainly high-voltage
distribution and transmission network). The fixed part of the tariff is recommended to be
differentiated with customers fuse size.”
Capacity pricing with AMS meters (“Prising av overføringskapasitet med AMS”)
(Report from THEMA, 2013).
18
http://www.thema.no/wp-content/uploads/2015/04/THEMA-rapport-2013-23-
Prising_av_overf%C3%B8ringskapasitet_med_AMS1.pdf
“The report analyzes the impact of AMS metering on the design of the distribution tariff
and to what extent prices should be used to signal operation costs and capacity shortages
in distribution networks. The report concludes that tariffs reflecting the hourly cost of
grid losses should be implemented, taking into account hourly electricity prices and
estimated loss ratios. Such grid pricing would enhance the precision of hourly spot price
signals and yield improved data on price response for end-users. The public good
characteristics of the grid (particularly that investments happen in leaps) make it
challenging to design precise capacity pricing without risking significant deadweight
losses. In addition, there are a number of practical obstacles to capacity pricing,
including lead times, security constraints and CENS (Cost of energy not supplied)
regulations.”
Recovery of residual costs with AMS meters (“Innkreving av residual nettkostnader
med AMS”) (Report from THEMA 2013).
http://www.thema.no/wp-content/uploads/2015/04/THEMA-rapport-2013-22-
Innkreving_av_residuale_nettkostnader_med_AMS.pdf
“The introduction of smart meters in the Norwegian distribution grid gives new
opportunities for network cost recovery through residual tariffs. Ideally, such tariffs
should be neutral both in the short and long run and not affect end-users demand for
electricity. The report finds that increased fixed charges and lower energy charges are
more efficient than the current high energy charges for households and other small end-
users, and may be introduced independently of smart meters. Introducing residual tariffs
based on maximum consumption for smaller end-users, which is made possible by hourly
metering, will also increase efficiency compared to the current model. However, there are
many administrative issues which must be sorted out before choosing a specific model.”
4.1.1 Sweden:
There have been some studies on the impacts of load tariffs on customer behaviour.
Among others, they have focused on how electricity use is impacted with new tariffs
based on capacity instead of energy consumed. This section presents some of the recent
Swedish research on the topic.
Bartush C, Wallin F, Odlare M, Vassileva I and Wester L, “Introducing a demand-
based electricity distribution tariff in the residential sector: Demand response and
customer perception”, Energy Policy 39, 2011.
“Increased demand response is essential to fully exploit the Swedish power system, which
in turn is an absolute prerequisite for meeting political goals related to energy efficiency
and climate change. Demand response programs are, nonetheless, still exceptional in the
residential sector of the Swedish electricity market, one contributory factor being lack of
knowledge about the extent of the potential gains. In light of these circumstances, this
empirical study set out with the intention of estimating the scope of households’ response
to, and assessing customers’ perception of, a demand-based time-of-use electricity
distribution tariff. The results show that households as a whole have a fairly high opinion
of the demand-based tariff and act on its intrinsic price signals by decreasing peak
demand in peak periods and shifting electricity use from peak to off-peak periods.”
19
C. Bartusch, P. Juslin, U. Persson-Fischier och J. Stenberg, ”Elkonsumenters
drivkrafter för en ökad förbrukningsflexibilitet. Hushållsattityder och anpassningar
till en tidsdifferentierad och effektbaserad elnätstariff”, Elforsk rapport 14:41, 2014.
http://www.elforsk.se/Rapporter/?rid=14_41_
“In terms of changes in the pattern of electricity use, the estimated effects of a demand-
based, time-of-use electricity distribution tariff are fairly marginal and limited to
households living in single-family houses. The results suggest that this category of
homeowners in Sollentuna, as part of adjusting their electricity consumption to the
demand-based tariff, has decreased their demand by 2.3 and 1.2 per cent during the
summer and winter months respectively as compared to the reference group in Saltsjö-
Boo…
The results further indicate that demand-based tariffs have an evident effect on
households’ attitudes and intentions to shift electricity use from peak- to off-peak hours,
but also that these are not reflected in their actual behavior. There is consequently only a
weak relation between the occurrence of a demand-based tariff and the share of
electricity that is consumed in peak- and off-peak hours respectively among the
households that the study covered. The psychological factors that have the most influence
on whether or not, and in that case to what extent, households adapt their electricity use
to a demand-based tariff in this sense are the expected consequences of, and so their
attitude towards, the behavioral change at hand as well as the degree to which they
perceive to have control over it. In more concrete terms, this means that economic
savings and positive effects on the environment, climate change and a sustainable
development for younger and future generations are the most important driving forces,
while the circumstance of considering ones electricity consumption to be basically non-
existing, or that one uses most of the electricity during off-peak hours as it is are the
greatest barriers, for adapting ones electricity consumption to the demand-based tariff.”
Einar Persson, Björn Berg, Fredrik Fernlund, Olle Lindbom, 12:48 Pilotstudie i
Vallentuna - Reflektioner rörande affärsmodeller för förbrukarflexibilitet och
självlärande prognosstyrning för kundanpassad effektreglering, Elforsk rapport 12:48,
2012.
http://www.elforsk.se/Rapporter/?rid=12_48_
”It is today possible to implement Demand Response in a large scale in Sweden through
smart services which automatically optimizes heating. For a homeowner with a ground
source heat pump this implies yearly savings of 2200-2600 SEK (electric boiler 2800-
4000 SEK), based on simulations of price and temperature data from the years 2010 and
2011. This is made possible through a combination of increased energy efficiency (10-15
%) due to effects such as smoother indoor temperatures, and shifting of consumption
from expensive to cheaper hours (approx. 15 kWh heat per day). The reward of load
shifting varies heavily between seasons, since it depends on an intricate relation between
price volatility and heating needs (e.g. 1720 SEK and 590 SEK for 2010 respectively
2011, simulated on ground source heat pump). It is also clear that the effect of time
dependent grid tariffs dominate the savings in times of low price volatility (e.g. of 1090
SEK “load shift savings” for an electric boiler during 2011 770 SEK was due to grid
tariffs). The benefits can be reached without demanding active participation or
compromising the comfort level.”
SWECO, ”Syntes av elnätstariffer”, Energiforsk rapport 2015:170.
20
“For some time, several parallel discussions regarding the use of the local distribution-
grids and necessary changes in the DSO-tariffs for low-voltage customers, have been
taking place…
Most likely, the optimal tariff-structure differs from grid to grid and between types of
customers. It is obvious that different customer types have varying preferences (and
potential to alter their consumption-pattern). So instead of ending in a discussion of
which tariff is most suitable for which customer, the cost-reflectiveness should be the aim.
Rather that speaking of “favouring” tariff-structures, the scope of discussion should
instead be to which extent the consumption-pattern is increasing the costs for the DSO.
This project have taken on the challenge in a novel fashion and started from real hourly
consumption data from almost 200 000 customers during a period of three years. The
project have “recalculated” the economic history and by this created new understanding
of how alternate tariff-models can affect costs/revenues, partly on an aggregated level
and partly for different types of customers. Both from the perspectives “ceteris paribus”
(everything else alike), that the change in the energy-system will accelerate with
increasing influx of electric cars and PV-systems, as well as that the technical
development facilitates automatic load control for a larger share of the customers.”
4.1.2 Finland:
Lassi Similä, Göran Koreneff, Veikko Kekkonen ”Network tariff structures in Smart
Grid environment.” RESEARCH REPORT VTT-R-03173-11, 2011.
http://www.vtt.fi/inf/julkaisut/muut/2011/VTT-R-03173-11.pdf
”The report examines the emerging options and requirements for electricity transmission
and distribution network tariffs in the evolving Smart Grid (SG) environment. Smart
Grids mean more sophisticated metering and communication technologies, which
enhance possibilities to make tariffs more transparent, more economically efficient and
cost-reflective, and more just.”
Jarmo Partanen, Samuli Honkapuro, Jussi Tuunanen, and Hanna Niemelä: ”Tariff
scheme options for distribution system operators” Lappeenranta University of
Technology,4.5.2012.
http://www.lut.fi/documents/10633/138922/Tariff+scheme+options+for+distribution+sys
tem+operators/d2c7a66f-4033-42ff-a581-dc4ef8586592
“The research report provides the results of the research project “Tariff scheme options
for distribution system operators”.
Kimmo Lummi; Antti Mäkinen, Antti Mutanen, Pertti Järventausta - Tampere
University of Technology ”Electricity Distribution Pricing Methodology In Finnish
Regulation Framework - A Case Study of Matching Principle”
Conference/Compilation NORDAC 2014, The 11th Nordic Electricity Distribution and
Management Conference, 8-9 September 2014, Stockholm, Sweden.
“In the paper, the practical theory of electricity distribution pricing in Finnish regulation
framework and electricity market environment is discussed. A methodology to produce
21
distribution tariffs, that realize the matching principle, is introduced. One goal of the
paper is to demonstrate the effects of different load profiles in tariff design.”
Energiavirasto (Energy Authority) ”Sähkön siirtohintatariffien kehitys 2000-2013
(Development of electricity network tariffs in 2000 – 2013)”, in Finnish.
http://www.energiavirasto.fi/documents/10179/0/Sahkon_siirtohintatariffienkehitys2013.
pdf/49f73b2d-f227-473f-b510-fb77a76f18e4
“The report examines development of fixed and variable components of electricity
network tariffs in Finland during period of 2000 – 2013.”
4.1.3 Denmark:
SEAS-NVE NET A/S og DONG Energy Eldistribution A/S ”Flyt dig! Forsøg med
variable nettariffer 2015 (Move! Experiment with variable tariffs)”.
http://www.seas-nve.dk/~/media/2-0-seas-nve/pdfer/om%20seas-
nve/15895%20rapport%20flyt%20dig_web.ashx
“The purpose of Move! was to find out to what extent ordinary households without
electric heating would change their behaviour as regards their consumption of electricity
when time-differentiated tariffs were used instead of fixed grid tariffs. The time-
differentiated grid tariffs were divided into three periods of the day on all days of the
week.
An analysis of the test data shows that the test customers reduced their consumption
during the peak load (red interval) by a little more than 2.0% in comparison with the
customers in the control group (equivalent to a change by 0.4 percentage points,
measured in relation to the total consumption). Furthermore, a test shows that this is a
statistically significant effect, and therefore the difference in the peak load share of the
electricity consumption between the test and the control groups can be attributed to the
Move!
An elasticity for the change in the peak load price compared to the total average
electricity price has been calculated at approximately - 0.19. This means that a change in
the relative price between peak load and the total average price of 10% will reduce the
consumption during peak load by 1.9%.
The effect with a reduction of the peak load share by 2% is relatively constant over time
and largely the same on weekdays as in the weekend.
The Move! test period was limited to one year. Thus, the effect uncovered by the
experiment must be described as a short-term effect. It is an open question whether the
effect will be of the same size in the long term.”
22
4.2 What can be learnt about design of tariffs from the survey of studies done?
The studies mentioned in chapter 4.1 have a wide range of suggestions for the future
implementation of tariffs. Overall the results from the studies can be arranged into the
following two groups:
4.2.1 Cost-reflective tariffs
A view from the studies is that tariffs should be cost-reflective in a way which reflects the
hourly cost of grid losses while taking into account electricity prices and estimated loss
ratios. Such grid pricing would enhance the precision of hourly spot price signals and
yield improved data on price response for end-users. One suggestion for a cost-reflective
design for load tariffs (based on the Norwegian network cost-structure) is to
approximately allocate the income from the different tariff components in the following
way: 20 percent from energy component, 10 percent from the part of the fixed component
that covers customer-specific costs, 20 percent from from the part of the fixed component
that covers costs mainly in the low-voltage network and 50 percent from from the part of
the fixed component that covers administrative costs and costs in the overall parts of the
network.
4.2.2 Demand response potential
Demand response has also been studied thouroughly and one of the studies shows that
households react fairly strongly to price signals by decreasing peak demand in the peak
periods and shifting electricity use from peak to off-peak periods. A study of time-of-use
electricity distribution tariff shows that the tariff has a fairly marginal and limited change
in the pattern of electricity use for households, the result is that homeowners (in
Sollentuna) have decreased their electricity demand by 2,3 and 1,2 percent during the
summer and winter months respectively. Antoher study found a similar result that time-
differentiated tariffs made customers decrease their consumption during peak loads with a
significantly 2 percent (in comparison with a control group).
Demand response can according to a study provide homeowners with a profit (2200-2600
SEK) by having their heating automatically optimized.
23
Annexes
Annex 1. Danmark (Denmark)
Annex 2. Finland (Finland)
Annex 3. Island (Iceland)
Annex 4. Norge (Norway)
Annex 5: Sverige (Sweden)
24
5 Tariff design in Denmark
The appendix is composed of four parts. The first part shows how the grid tariffs are
treated in the Danish Electricity Act and how the Energy efficiency directive is
implemented in the Act. The second path gives an overview of the tariff design in
Denmark and an actual picture of the structure of load tariffs for the DSO-grids. The third
part is a status for smart meters in Denmark. Finally, the fourth part is a short survey over
the discussions of tariff design.
Analysing design of grid tariffs is important for two reasons:
New technologies such as interval metering and two-way communication
provide new opportunities for tariff structures.
Different tariff structures give consumers different incentives and can
therefore impact the use of the grid
New technologies like metering and communication gives opportunities for a more
flexible use of electricity over the hours during a year. New technologies can therefore
make the customers more price sensitive in the future.1 The new metering device can
change the design for household customers and small firms from traditionally energy
based to power based charging.
5.1 Few restrictions in how to design grid tariffs in Electricity Act
Grid tariffs are treated in Chapter 10 of the Danish Electricity Act, which places the
following requirements on grid tariffs:
§ 70: Prices for grid companies’s services must be set in accordance with the revenue cap,
which are set to cover cost for an efficiently operated grid.
§ 73: Grid tariffs must be set in a fair, objective and non-discriminatory way, according to
the costs that the respective consumer groups apply. Price differentiation for the sake of
an effective use of the grid and security of supply are legal (this last part apply from 1. of
April 2016).
§ 73 a: The (tariff) methods used by the grid companies must be approved by the Danish
Energy Regulatory Authority (DERA) before use.
§ 73 b: The industry organisations can make standardized guidances for methods of tariff
setting etc., which DERA supervise.
On top of it is stated in the first paragraph of the act that the overall purpose of the
electricity act amongst others are to secure an effective use of the economic resources,
and in § 6 it is stated, that the grid companies must offer their services to the consumers
in an transparent, objective, fair and non-discriminatory way.
1 Most studies of customers price elasticity reports rather low levels of elasticities.
25
Summarizing, the Electricity Act only puts few requirements on grid tariffs: be
transparent, fair, objective and non-discriminatory. As long as these requirements are met,
grid companies are free to set grid tariffs as they see fit. DERA sets the size of the total
allowed revenue (revenue cap), but does not decide how grid tariffs are structured –
DERA, however, approves the methods used by the DSO for setting tariffs.
The Energy efficiency directive is implemented in the Danish electricity act – most of the
directives content was already implemented in advance of the passing of the directive,
and the directive has not caused any changes in the electricity act regarding tariff setting.
5.2 Structure of grid tariffs for load in DSO-grids
DERA collects the tariffs from each DSO in Denmark. In Denmark it is required that the
methods used by the DSO for setting tariffs are approved by DERA, and the DSOs are
obliged to report to DERA in case of a change in tariffs.
The methods for tariff calculation must comply with the Danish electricity Act, that states
that the methods must ensure that the tariff are set in a fair, objective and non-
discriminatory manor according to the cost that a customer group apply. This also means
that there must be no cross subsidization amongst customer groups.
The DSO can set their methods for tariff calculation based on an industry guidance,
which DERA supervise. The industry guidance are devised by the Danish Energy Agency
(DEA) and supervised by DERA. Most of the DSOs use the industry guidance with minor
deviations.
The cost structure in net distribution are related to tariffs by principles of cost allocation.
The cost allocation has the purpose of fulfilling the goals for objective (cost-reflective)
tariffs. The different type of costs are allocated to the different customer groups, which
are set by their connection point (voltage level) to the grid, and thereafter it is decided
whether the different types of costs should be tariffed as a part of the fixed tariff or the
variable tariff, and whether it should be tariffed as a downward running tariff and should
be time diffentiated.
The industry guidance contains unambiguous guidance concerning all of these elements
in the tariff methods. The overall guideline is, that all cost types for the DSO that depend
on the consumption of electricity are tariffed as a part of the variable tariff and all cost
types that do not vary with consumption (i.e. costs concerning administration and costs
concerning meters and metering) are tariffed as a part of the fixed tariff. On top of that
most of the variable costs are in the just supervised industry guidance recommended to be
time differentiated. This industry guidance is expected to be used by most of the DSOs by
April 2016.
This means, that there are no or very little variation in the design of the tariff structure
amongst the Danish DSOs. However, this does not mean that the fixed and the variable
part of the tariffs constitute the same part of the total tariffs amongst the DSO, since this
is determined by the cost structure of the respective DSO.
In Figure 3the proportion of the fixed part in the tariff for the following customers: A
customer connected on the low voltage level with a consumption of 4000 kWh electricity;
a customer on the middle voltage level (or in the lower side of a transformer) with a
consumption of 1 mio. kWh; a customer on low voltage level with a consumption of 1
26
mio. kWh and finally a customer on the high voltage level with a consumption of 25 mio.
kWh.
For the small customers the fixed part varies from 0 to 74 % with far most between 34
and 54 %. For the larger customers the fixed part is below 10 % for most grid companies.
Figure 3 | Fixed part of a grid tariff for different levels of consumption, 2015
0%
10%
20%
30%
40%
50%
60%
70%
80%
1 11 21 31 41 51 61
Distribution system operator
4.000 kWh 1.000.000 kWh (Middle voltage)
1.000.000 kWh (Low voltage) 25.000.000 kWh
Source: DERA
As can be seen in Figure 3, the fixed part of the tariffs varies quite much especially
between the DSOs for the small costumers – for larger costumers the variations are
smaller. For the small customers the fixed part varies from 0 to 74 %, but with far most of
the DSOs with fixed parts between 34 and 54 %. For the larger customers the fixed part is
below 10 % for far most of the DSOs.
Table 3 shows the average variable tariffs and average fixed part for year 2015.
Table 3 | Mean tariffs for different customer groups, 2015
Variable tariff (øre/kWh) Fixed tariff (DKK/year) Household consumer (C-customer on low voltage)
20,05 525
Industry consumer (low voltage)
12,95 2.800
Industry consumer (B-customer on middle voltage)
9,70 5.400
Large industry consumer (A-customer on high voltage)
4,01 17.500
Source: DERA
Table 3 shows that the higher voltage level that a costumer is connected to the grid, the
lower variable tariff (caused by the structure with downward running tariffs), but the
higher fixed tariff, caused by higher administrations costs, but due to the high
consumptions of the costumers on higher voltage levels, the share of the fixed tariff in the
overall tariff payments are still lower, than for small costumers.
27
In Denmark tariffs only account for a minor share of the total electricity price, cf. Figure
4 for an overview of the different shares of the electricity price.
Figure 4 | Elements of the electricity price in 2015
Source: DERA Note: *Public Service Obligation
Figure 4 shows that tariff only accounts for a minor part of the total electricity price – in
particular for small costumers, for whom the tariffs accounts for only about 15 pct. of the
total price. This is due to the high electricity taxes in Denmark. For larger scale costumers
(business/industry consumers) the tariffs account for a larger share (about 27 pct) due to
the fact that Industry consumers do not pay VAT and only pay a very small energy tax.
5.3 Smart meters
By the end of 2013 it was decided to implement AMR-systems (smart meters) to all
consumers in Denmark. At the time only around 60 percent of the consumers had a smart
meter, including industrial consumers with a consumption of more than 100.000
kWh/year, whom are obliged to be billed by the hour, and thereby had to have a smart
meter. This means that the new rules mainly apply to template customers whom has a
consumption of less than 100.000 kWh/year.
The AMR-systems shall be fully implemented by 2020. Figure 5 below shows the status
for the rolling out of smart meters for small consumers by the end of 2014, illustrated by
the already installed and planed installations in 2014 by the end of 2013. As shown, the
status by the end of 2014 was, that 57 pct. of the small consumers should have a smart
meter. Figure 5 also shows, that it is mainly in the Copenhagen area, that the
implementation is in the starting phase.
28
Figure 5 | Implementation of smart meters in Denmark, 2013
Source: Danish Energy Agency
5.4 Discussions on tariff design in industry
With the continuous implementation of AMR systems for reading the customers use of
electricity it is now possible to use power based charging hourly for household and other
small users of electricity. Therefore this has been an issue in the industry regarding the
design of grid tariffs.
The industry Association, Danish Energy Association, has in the new industry guidance,
just approved by DERA, developed a model for time-of-use tariffs. The model is a
simplified model with the purpose to give the DSOs the opportunities to test time-of-use
tariffs and collect experiences rarther than a theoretical correct cost reflective model. The
simplified model is also ment to be a precursor for this correct cost reflective model.
Small consumers – smart meters installed or
planned to be in 2014. Assessed ultimo 2013
Aggregated approx. 57 pct.
29
The time differentiation is based on standardized load curves on daily basis on aggregated
level for all DSOs, and separated between the higher voltage levels and the lower voltage
level and between summer and winter. Load curves are calculated as the average load in
the grid hour by hour. The idea is that extra costs due to high load in the grid are paid by
consumption during these high load periods. This simplified model is developed to
minimize the respective DSOs need to get further information for tariffing. In Figure 6 a
fictive daily curve is illustrated with the limits for time differentiation.
Figure 6 | Daily load and limits for time differentiation, fictive example
Tid
Kl. 0 Kl. 24
Døgnbelastning, pct.
65
90
Højlast
Spidslast
Lavlast
Source: The DERA Secretariat. Note: The figure is an illustration of a fictive daily load curve.
In Figure 7 the time differentiation of the tariffs in the simplified model are shown. A
green dot means low load, a yellow dot means high load and a red dot means peak load.
In the calculation of the tariffs the idea is, that consumption in high load periods shall
contribute to the share of costs allocated to low load. In periods with low load you only
pay the cost for this low load. In periods with high load you pay the cost of low load and
an additional charge for the cost of consumption in the high load period, and in periods
with peak low you pay the cost for low load, plus additional charge to high load and again
an additional charge for peak load consumption.
According to Dansk Energi the starting point is that by exploiting demand response
flexibility it is possible to postpone or avoid development of the network (increase in the
grid volume). Therefore it is more costs over time and not specific marginalcosts at a high
consumption for a given hour. The time differentiated tariffs sends the right pricesignals
in a long term perspective, because it reflects the costs of consumption under the maximal
peak load. By using time differentiated tariffs the consumers are incentivized to utilize
the grids capacity in an appropriate manner.
Daily load, in percent
Peak load
High load
Low load
Hour of
the day 00:00 24:00
30
Figure 7 | Timedifferentiation of tariffs, fictive examples
Source: Dansk Energi. Note: Green means low, yellow high and red peak.
5.5 Conclusion
The efficiency directive has been implemented in the Danish electricity act, and the tariffs
must be set in a transperant, fair, objective and non-discriminatory manner. The tariff
structure in Denmark contains a fixed part covering all non-comsumption dependend
cost, and a variable part covering the cost that depends on the consumption. The fixed
part of the tariff payment varies between the DSOs, from zero to more than 70 percent.
The industry has recommended a simplified model for time-of-use tariffs which are
expected to be implemented in the DSOs from April 2016, but not for all consumers from
that point, since AMR-systems will not be fully implemented until 2020.
31
6 Tariff design in Finland
6.1 Electricity Market Act
The new Electricity Market Act (588/2013) defines certain standards and principles that
pricing of electricity distribution services must meet. These are covered mainly by § 24
and § 25 of the Electricity Market Act.
According to § 24 subsection 1, the sale prices and terms of network services and the
criteria according to which they are determined shall be equitable and non-discriminatory
to all network users. Exceptions to them may only be made on special grounds.
This provision means that the network operator must provide network services by same
uniform terms for all the network operator’s customers. Distribution pricing must not
depend on the customer’s supplier of electricity. Pricing must not change unfoundedly
when the electricity supplier changes. Pricing must be uniform within same customer type
group. Pricing can vary between different size and type of customers.
According to § 24 subsection 2, pricing of network services has to be reasonable.
This provision means that network tariffs have to reflect costs that the network operator
has possibility to achieve. Energy Authority regulates network operator’s total revenue
from the network services. Regulatory methods ensure reasonable return for capital
committed to network operations and incentivize cost efficiency.
The Electricity Market Act 24 § subsection 3 requires that the pricing of network services
shall not contain any conditions or limitations that would be unfounded or that would
obviously restrict competition within electricity trade.
The aim of this provision is to ensure that the terms of network services are reasonable
and do not unfoundedly restrict the customers' ability to utilize the network.
According to Electricity Market Act, § 25 subsection 2, so-called spot pricing is applied
in pricing of electricity distribution services.
The network operator shall, for its own part, create preconditions permitting the customer
to conclude a contract on all network services with the network operator to whose
network he is connected as subscriber. The network operator shall, for its part, create
preconditions permitting the customer to be granted the rights, in return for payment of
the appropriate fees, to use from its connection point the electricity system of the entire
country, foreign connections excluded.
Within a distribution system, the price of system services must not depend on where
within the network operator's area of responsibility the customer is located
geographically.
According to Electricity Market Act 54§ distribution network operator must offer
different network services needed by users of its network.
32
The network operator must provide end users time-based electricity distribution services.
This includes time-based metering service. Time-based electricity distribution services
include: distribution based on hourly metering, distribution based on one time electricity,
distribution based on two-time electricity (day- and night time) and distribution based on
seasonal electricity (winter day time and other).
The Electricity Market Act was amended in 2014 (1430/2014). According to § 24 a
pricing and terms of sale of network services shall not contain criteria that are harmful for
overall efficiency and energy efficiency of production, transmission, distribution and
delivery of electricity.
In addition to obligations imposed directly by the law tariffs must meet the following
more general principles:
Principles:
Tariffs must be cost-reflective
Tariffs must be impartial and non-discriminatory for different customer groups. In
addition, customer can choose any tariff product offered at the voltage level in
question.
Tariffs must be simple and clear to understand for customers.
Tariffs must also steer customers to efficient use of the power system and thus
encourage customers to save energy.
The objectives imposed to tariffs are partly contradictory. It is necessary to seek a balance
between different objectives when determining the tariffs. Cost reflectivity of tariffs will
actualize on average when the cost reflectivity of tariff products provided by the network
operator is optimized. Large fixed part in tariff may not encourage consumers to save
energy, but it may be more equal when allocating costs.
There are no regulations on how the network operators should allocate costs on the fixed
and variable parts of network tariff. More detailed definition has been left to
consideration of the electricity network operator.
6.2 Structure of grid tariffs in DSO-grids Distribution system operators (DSO) have obligation to report electricity distribution grid
tariffs in effect to Energy Authority. Based on reported tariffs Energy Authority has
calculated average tariffs for different user types2. The user types represent typical
households, with and without electric heating. There is also a user type for small
industrial user.
User types are:
2 For more information, see: Energiavirasto (Energy Authority) ” Sähkön siirtohintatariffien
kehitys 2000-2013 (Development of electricity network tariffs in 2000 – 2013)”, in Finnish
http://www.energiavirasto.fi/documents/10179/0/Sahkon_siirtohintatariffienkehitys2013.pdf/49f73
b2d-f227-473f-b510-fb77a76f18e4
33
Apartment, no electric sauna heater, main fuse 1×25 A, consumption 2000 kWh /
year
House, no electric heating, electric sauna heater, main fuse 3×25 A, consumption
5000 kWh / year
House, electric heating, main fuse 3×25 A, consumption 18000 kWh / year
House, electric heating, main fuse 3×25 A, consumption 20000 kWh / year
Small industry, power requirement 75 kW, consumption 150000 kWh / year
Fixed component of grid tariff consist of fixed standing charge, usually depending on the
fuse size, and with the industrial users, possibly additional power fee. Possible metering
equipment fee is included in the fixed component of the tariff. Variable component of
tariff is based on consumption of electricity.
Fixed component of grid tariff varies a lot among DSOs. Figure 8 shows that for a user
type with consumption of 2000 kWh/year, fixed component of tariff varies from 30% to
over 80% depending on which DSOs area consumer is located. Fixed component’s share
of grid tariff is usually higher on DSOs operating in rural areas than DSOs operating in
suburban or urban areas.
Figure 8 | Fixed part of the grid tariff for user type with 1×25 A, 2000 kWh/year in 2015
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
1 11 21 31 41 51 61 71
Distribution system operator
Source: DERA, Energiavirasto, NVE & Swedish Energi Market Inspectorate
Table 4 shows the average fixed part, variable part and total grid tariff of above
mentioned user types, not including taxes. From January 2010 to January 2015 the
average grid tariff seems to have decreased in all consumer type categories except
apartments consuming 2000 kWh/year.
34
Table 4 | Average tariffs by user type (in year 2015 price level, not including taxes)
kWh/year Fixed part € Variable part € Total €
2010 2013 2015 2010 2013 2015 2010 2013 2015 Apartment 2.000 (1×25 A) 49 61 63 52 54 50 101 115 112
House 5.000 (3×25 A) 94 97 99 144 135 125 238 232 224
House (electric heating)
18.000 (3×25 A) 181 187 188 378 358 337 559 545 525
House (electric heating)
20.000 (3×25 A) 180 187 188 353 336 317 533 523 505
Small industry
150.000 (75 kW) 1.372 1.669 1.700 3.819 2.828 2.520 5.191 4.497 4.221
Source: Energiavirasto
Table 5 illustrates how relative share of fixed and variable tariff components has
developed over the years. The share of fixed component has increased in all consumer
type categories.
Table 5 | Proportion of fixed and variable parts of network tariff by user type
kWh/year Fixed part % Variable part %
1/2010 1/2013 1/2015 1/2010 1/2013 1/2015
Apartment 2000 (1×25 A) 49 % 53 % 56 % 51 % 47 % 44 %
House 5000 (3×25 A) 40 % 42 % 44 % 60 % 58 % 56 %
House (electric heating) 18000 (3×25 A) 32 % 34 % 36 % 68 % 66 % 64 %
House (electric heating) 20000 (3×25 A) 34 % 36 % 37 % 66 % 64 % 63 %
Small industry 150000 (75 kW) 26 % 37 % 40 % 74 % 63 % 60 %
Source: Energiavirasto
Figure 9 and Figure 10 show the development of components forming price of electricity.
Share of energy component has decreased over the years in consumer type categories of
apartment and house. There has been significant increase in the share of distribution tax
component from January 2010 to January 2015.
35
Figure 9 | Components of the retail electricity price, house with electric heating - 2.000 kWh/year
0%
10%
20%
30%
40%
50%
2010 2011 2012 2013 2014 2015Energy (1×25 A, 2.000 kWh/year) Distribution (1×25 A, 2.000 kWh/year)
Tax (energy) (1×25 A, 2.000 kWh/year) Tax (distribution) (1×25 A, 2.000 kWh/year)
Source: Energiavirasto
Figure 10 | Components of the retail electricity price, house with electric heating - 18.000 kWh/year
0%
10%
20%
30%
40%
50%
60%
2010 2011 2012 2013 2014 2015Energy (3×25 A, 18.000 kWh/year) Distribution (3×25 A, 18.000 kWh/year)
Tax (energy) (3×25 A, 18.000 kWh/year) Tax (distribution) (3×25 A, 18.000 kWh/year)
Source: Energiavirasto
According to the government decree on determination of electricity supply and metering
(Vna 66/2009)3, the metering of electricity consumption and small-scale electricity
generation must be based on hourly metering and remote reading of the metering
equipment (hourly metering obligation). Hourly metering must cover at least 80 % of the
DSO’s electricity metering points by the end of 2013. Figure 11 shows that by the end of
2014 89% of installed smart meters were at use of balance settlement.
3 Valtioneuvoston asetus sähköntoimitusten selvityksestä ja mittauksesta (Government decree on
determination of electricity supply and metering), (VNa 66/2009)
36
Figure 11 | Proportion of installed smart meters
38% 39%
56%
79%
95%
12%
19% 20%
52%
89%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2010 2011 2012 2013 2014
Smart meters installed Smart meters at use in balance settlement
Source: Energiavirasto
6.3 Discussions on tariff design
In 2012 a research group of LUT Energy conducted a research project “Tariff scheme
options for distribution system operators”. The project steering group comprised the LUT
Energy researchers and representatives from The Finnish Energy Industries, electrcity
network operators, Energy Authority and Ministry of Employment and the Economy.
The project report concluded that new pricing schemes are needed to encourage the
customers in energy efficiency and demand response. According to the report the target is
to establish a pricing scheme for DSOs that encourages the end-users to behave so that
the energy efficiency of the whole energy system, including generation, transmission and
distribution, is maximised and the total costs to the national economy are minimised.
Furthermore, the pricing scheme has to be cost reflective, equitable and intelligible to all
parties involved. Report studies the following alternative pricing schemes; fixed monthly
charge, energy rate, dynamic energy tariff, power-based tariff and takes a closer look on
power band pricing scheme.
In 2013 Energy Authority sent to selected DSOs a questionnaire about objectives of their
tariff policy and pricing. In the answers, DSOs give reasons for their selection of tariffs,
which are: fulfilling authorities’ recommendations on means of metering, taking into
account different customer needs and providing enough choices for consumers of
different size. Most common objectives of tariff policy are steering and different
objectives related to customers. The aim is to steer customers to efficient and stable use
of electricity, and to save energy.
Customer related tariff policy objectives are for example clarity of tariff structure,
selection of tariff products and ease of changing the product. Ensuring continuity of
DSO’s operations, predictability of revenue and stable income for owners is seen
important.
37
Cost structure and the cost of network operations are mentioned as the basis of tariff
setting. Also, Energy Authority’s regulation methods and financing needs of replacement
investments are mentioned. In practice, the network is designed and measured based on
peak load capacity. According to some answers this has direct cost effect on construction
and maintenance of network. According to some DSOs, current structure of their
distribution tariffs do not reflect their true cost structure. In this case, fixed costs are
allocated to some extent also to energy consumption based (variable) tariff components.
The fixed standing charge (fixed tariff component) would increase significantly if all the
fixed costs would be allocated on them. According to the DSOs, the actual share of fixed
costs is 75% – 90%.
38
7 Tariff design in Iceland
7.1 Electricity Act In Iceland each DSO shall establish a tariff for its services in accordance with the income
possibility curve decided every five years by the National Energy Authority Orkustofnun
(NEA).
The main provisions on tariffs in DSOs grids are to be found in Article 17 a. of the
Electricity Act no 65/2003.
The same tariff shall apply in the distribution zone of each DSO for the consumption of
low voltage electricity, i.e. 230-240 V. If the energy from the distribution system is
delivered at a different voltage the tariff may be adjusted accordingly. In the same way,
account may be taken of other differences in service in the tariff.
DSOs are permitted to apply to NEA for permission to maintain a separate tariff for rural
areas where the cost of distribution is demonstrably higher than in urban areas. The
condition for permission to maintain a separate rural tariff is that the use in the rural area
in question must amount to a minimum of 5% of the total use in the distribution zone of
the distribution system operator.
Each DSO shall send the tariff to NEA two months before taking effect. If NEA is of the
opinion that the submitted tariff is in violation of law, NEA shall submit its comments to
the distribution system operator in question within two weeks of the submission. A tariff
shall not take effect until rectifications have been made to the satisfaction of NEA. The
DSO is then obliged to publish the tariff.
7.2 Structure of grid tariffs in DSO- grids The tariff for households consists mainly of two components, a fixed component and
component based on the household´s use of electricity, c.f. Figure 12. Furthermore the
customer pays an energy tax on the distribution cost and VAT. The tariff is the same
during all hours of the year.
Figure 12 | Components of electricity costs to a household, 2015
30%
49%
21%Electricity
Tariffs
Taxes
Source: Orkustofnun
39
The rules on tariffs in Iceland grant the DSO a wide margin of discretion to decide the
tariffs given that they treat their customers equally and establish their tariffs within the
limits of their revenue caps. Therefore the tariffs for the six operating DSOs in Iceland
are different.
Due to geographical limitations the cost of distribution to users in some areas of the
country can be higher than in other areas. As mentioned above, DSOs can apply to NEA
for a permission to maintain a separate tariff for rural areas if the cost of distribution is
demonstrably higher than in urban areas. Currently, two DSOs have been granted such a
permission.
7.3 Discussion on tariff design The Icelandic DSOs have not implemented smart meters for their customers and NVE has
not made any recommendations to the DSOs on that issue. Since the tariffs for DSOs in
Iceland are the same during all hours there is a lack of incentive in implementing smart
meters.
7.4 Conclusion Iceland has not implemented the Energy Efficiency Directive since the Directive has not
yet been made part of the Agreement on the European Economic Area. At the moment
the government does not have any changes on the agenda as regards tariff design for the
Icelandic DSOs. The implementation of the Directive may, however, have some impact
on the legal framework that the DSOs currently work under in the future and indirectly
effect tariff design in Iceland.
40
8 Tariff design in Norway
In the Norwegian regulations all tariffs are based on the costs referring to the consumer’s
point of connection. An agreement with the network company in the point of connection
shall provide access to the entire network system and the power market.
All network companies are responsible for framing tariffs within their revenue cap
pursuant to the regulation on tariff structure.
8.1 Electricity Act Grid tariffs are treated in The Energy Act Regulations chapter 4, which states that tariffs
are set by the network companies. Tariffs are used as the common term for all prices and
financial remuneration for connection and use of network installations. Tariffs shall be
designed so that they give signals whenever possible about efficient utilisation and
efficient development of the grid. Tariffs can be differentiated according to objective and
verifiable criteria based on relevant grid conditions.
Additional requirements are given in Regulations governing financial and technical
reporting, income caps for network operators and tariffs (Reg. No. 302 of 11 March
1999), part V:
Chapter 13. General provision on tariffs
Section 13-1. Principles of point-based tariffing
All network companies are responsible for working out point-based tariffs that are in
accordance with the following principles:
a) the tariffs shall refer to the points of connection.
b) an agreement with the network company in the point of connection shall provide
access to the entire network system and the power market.
c) the network company is obliged to offer all customers who want network services
non-discriminatory and objective point-of-connection tariffs and conditions.
d) the tariffs shall be designed so as to as far as possible give signals about efficient
utilisation and efficient development of the network.
e) the tariffs can be differentiated according to objective and verifiable criteria based on
relevant network conditions.
f) the tariffs shall be determined independently of power purchase agreements.
g) the tariffs shall provide the network company with income to cover costs under its
revenue cap, tariff costs for access to higher voltage network (superjacent network),
property tax paid and statutory payments to the energy fund.
h) all houses, apartments or vacation homes shall be metered and settled individually.
Section 13-2. General rules on tariffing
41
The tariffs for drawing and feeding power shall be worked out in accordance with the
following basic structure:
a) usage-dependant tariff components that vary with the customer’s ongoing drawing or
feeding of energy.
b) other tariff components.
Section 13-3. Usage-dependent tariff components
The usage-dependent tariff components consist of an energy component and a capacity
component.
As a main rule, the energy component shall be set on the basis of the cost of marginal
network losses.
The network companies may set a capacity component to create balance between
transmission needs and network capacity. The capacity component may be used when
transmission needs exceed the capacity in the network.
Section 13-4. Other tariff components
Other tariff components shall cover network costs that are not covered by usage
dependent tariff components.
Chapter 14. Practical design of tariffs for the ordinary drawing of power
Section 14-2. Design of tariffs for the ordinary drawing of power from the distribution
network
On the distribution network, customers without maximum demand metering shall be
charged a fixed component and an energy component such that:
a) the fixed component covers customer-specific costs and a share of the other fixed
costs on the network.
b) the energy component covers the cost of marginal network losses and may in addition
cover a share of the other costs not covered by the fixed component.
Customers with maximum demand (load) metering on the distribution network shall be
charged a fixed component, energy component and a load component. The fixed
component shall as a minimum cover customer-specific costs. The energy component
shall as a minimum cover the cost of marginal network losses. The load component shall
be based on the customer’s power consumption during defined periods.
Separate tariffs shall be prepared for high-voltage and low-voltage drawing.
For low-voltage drawing, the load component shall be volume-differentiated. These
tariffs shall be designed such that all customers pay the same price for drawing up to the
first stage and lower rates at subsequent stages. Tariffs may also be determined by other
means that yield the same result.
The network owner shall offer tariffs with a time differentiated energy component to all
customers in the distribution network who by regulation are subject to mandatory meter
readings several times per year.
42
8.2 Structure of grid tariffs in DSO-grids Current regulation provide network companies large degree of freedom regarding how to
design tariffs. Tariffs for households, vacation homes and small commercial customers
mainly consists of a fixed component and an energy component. Customers with master
fuse exceeding a set limit, for example over 80 Ampéres or 125 Ampéres at 230 Volt, or
customers with expected annual consumption exceeding 100 000 kWh usually have a
load component in addition to the fixed component and the energy component.
Network companies part their customers into groups offered different tariffs, based on
relevant network conditions. It is not irregular for households, vacation homes and small
commercial customers to have different level of the fixed part within the same network
company.
In Figure 13the proportion of the fixed part in the tariff for a household customer with 20
MWh per year is shown for 2015. The fixed part varies from 11% to 71% with mean
value of 35%.
Figure 13 | Fixed part of grid tariff for household customer with 20 MWh per year in 2015
0%
10%
20%
30%
40%
50%
60%
70%
80%
1 11 21 31 41 51 61 71 81 91 101 111 121
Dristribution system operator Source: NVE
In Table 6 mean tariffs for year 2015 in different customer groups is shown:
Table 6 | Mean tariff for different customer groups in 2015
Power Energy
kWh/year
Tariff
øre/kWh
(EUR/Mwh)
Vacation homes - 4000 82,5 (74,2)
Households
(weighted average - 20 000 26,3 (23,6)
Households - 20 000 33,2 (29,9)
43
Small industy - 30 000 35,6 (29,3)
Industry 40 kW 160 000 27,2 (24,4)
Industry 400 kW 1,6 GWh 11,9 (10,7) Source: NVE
In Table 7 tariffs for year 2011 and 2015 are shown regarding the proportion of fixed part
of total tariff.
Table 7 | Proportion fixed charge in grid tariff for different customer groups
Power Energy kWh/year Fixed part of tariff
2011 2015
Vacation homes - 4000 71 % 72 %
Households
(weighted average) - 20 000 30 % 31 %
Households - 20 000 32 % 34 %
Small industy - 30 000 27 % 34 %
Industry 40 kW 160 000 17 % 19 %
Industry 400 kW 1,6 GWh 1 % 1 % Source: NVE
Roughly, network tariffs constitute about 1/3 of electricity costs to a household, while
about 1/3 is payment for the electricity to the supplier and about 1/3 is fees to the state4,
cf. Figure 14. Fees consist of consumption tax, statutory payments to Energy Fund and
costs for electricity certificates.5.
Figure 14 | Components of electricity costs to a household, 2015
32,5%
26 øre/kWh
32,9%
26,3 øre/kWh
34,6%
27,7 øre/kWh
Electricity
Tariffs
Taxes
Source: Statistisk Sentralbyrå
8.3 Discussions on tariff design Within 1 January 2019, all Norwegian electricity consumers should have new advanced
metering systems (AMS). AMS-meters will measure the customers electricity
4 There are special arrangements in the three most northern counties, Finmark, Troms and
Nordland. Households in these areas pay no VAT, there is also no consumtion tax in Finmarks and
some of the minucipals in Troms. These areas will have a different mix, naturally. 5 Consumption tax was 0,1365 NOK/kWh in the first half-year of 2015. From 1.7.2015
consumption tax is 0,1415 NOK/kWh. Payment to Energy Fund is 0,01 NOK/kWh and expected
costs in 2015 for Electricity Certificates are 0,017- 0,021 NOK/kWh.
44
consumption on an hourly basis, and provide far better information about actual customer
usage. This was a decision made by NVE in 2011 and NVE defined features of these
meters.
NVE has collected information from most of the DSOs about how many customers have
advanced metering systems installed. By the end of 2014, 10 % of all customers had
advanced metering system with two way communication. (The majority of these are
installed before 2011 and do not satisfy all demands for AMS, but the companies must
not change these before 2019. In a setting where we compare the Nordic countries, 10 %
will be the reasonable share).
In addition to more accurate settlement of electricity and network tariffs, AMS enables
customers that adapt their consumption to price signals from the electricity market and
from network tariffs to reduce their costs relating to energy consumption. By measuring
the energy consumption not only in volume, but also over time, consumers may
contribute not only to more efficient utilisation of the grid, but also with flexibility
(demand-response) that may delay or reduce the need for grid investments. This will
benefit all customers through lower tariffs than in the case with grid investments.
Costs arising in the grid by consumer use are mainly related to losses, provided sufficient
grid capacity. The energy component of the tariff is far higher than the value of marginal
losses6. This affects the not only the cost allocation among customers, but also the
relative ratio between electricity from the grid and other options, without regard to the
power system. For example:
The prices between electricity and other energy carriers. The level of the energy
component affects the relative profitability of electricity versus fuel oil, gas or other
options in the choice of heating solution.
Profitability of energy efficiency measures, and profitability for electricity production
behind own meter (prosumer). For prosumers the energy component of the tariff is
netted out in the point of connection. This means that investment in energy efficiency
measures and production behind own meter can save contribution to cover capital and
operating costs.
The ongoing discussion regarding tariff design suggests less energy-based and more
power based tariffs for customers in DSO grids. NVE has had of a hearing on the issue,
and the industry organization “Energi Norge” has a working group concerning tariff
design. See also chapter 4 for some of the reports that have been written on the subject.
An introduction of power tariffs to all customers could be justified by the following:
better utilization of the grid, less grid investments and lower tariffs
more cost-reflective and fair allocation of grid costs among users of the grid since
network costs are far more capacity driven than energy driven.
NVE have heard the concept about the following tariff design for all load customers in
DSO grids:
6 Based on average historical costs, marginal losses will be about 0,05 NOK/kWh, while mean
value of the energy component in 2015 is 0,181 NOK/kWh.
45
An energy component reflecting the cost of marginal losses
A fixed component covering customer-specific costs
A power-based component, following one of the three options:
1. Customers are settled according to metered load (kWh/h) in a given hour e.g.
highest monthly hourly load or customers load during peak hour. One hour, or
mean value of several hours.
2. The customer pay a certain amount according to the main fuse size of the electrical
installation, as a measure of the customers potential power use from the grid.
3. The customer subscribes to a certain amount of maximal hourly power, and pays
according to this. For consumption above this amount, there is a higher price, or
it might be possible to limit the hourly power consumption to the subscribed
amount using technology in the AMS meter.
Power tariffs can be designed to affect consumption (capacity tariff) or to cover the total
need for income in the grid. The ideal way to cover residual costs, is a fully independent
neutral tariff which does not affect the use of the grid. A fixed fee that is equal for all
customers is the closest we get to such a residual tariff. When discussing tariff design, it
is therefore crucial to be clear about whether we want the tariff to give signals about the
load situation in the grid or not.
Shortage of grid capacity in the distribution grids is often solved by new grid investments,
although there are only few hours during the year where the demand for grid capacity
exceed the capacity in the grid.
NVE have heard the concept about handling shortage of grid capacity by utilizing
consumer flexibility.
End user flexibility can be understood in two different ways:
On the one hand, it may on the one hand be understood as price response, as end
users voluntary adaptation of own energy consumption to the current price signals
from the electricity market and grid costs.
On the other hand, end user flexibility may be understood as interruption of
consumption. Interruption is remote controlled and can in principle be carried out by
a grid company, an aggregator or another actor that the customer has an agreement
with. The end user must be compensated to accept that his consumption may be
interrupted. If one or more participants are willing to pay, and end users are willing to
sell there may occur a market for interruption, or a market for flexibility.
Purchase of end user flexibility in a market solution will provide proper valuation of
flexibility and visualize the opportunity cost of an investment. This is in contrast to the
current alternatives to grid investments as capacity pricing through tariff does not appear
as a cost to the network company, but only a redistribution of income.
NVE has received 57 letters commenting on the hearing, mostly positive to introducing a
power-based component in the tariff design. NVE will evaluate the comments and
46
consider whether there is need to change the regulations of the tariff design. This work
will continue in 2016.
8.4 Conclusion The Norwegian acts states that tariffs shall be designed so that they give signals whenever
possible about efficient utilisation and efficient development of the grid. Tariffs can be
differentiated according to objective and verifiable criteria based on relevant grid
conditions.
Current regulation provide network companies large degree of freedom regarding how to
design tariffs. Tariffs for households, vacation homes and small commercial customers
mainly consists of a fixed component and an energy component.
The ongoing discussion regarding tariff design suggests less energy-based and more
power based tariffs for customers in DSO grids. NVE has had a hearing on introducing a
power-based component in the tariff. This work continues in 2016.
47
9 Tariff design in Sweden This appendix is composed of four parts. An overview of the design of tariffs in Sweden
is followed by an actual picture of the structure of load tariffs for the DSO-grids. The
third part consists of a description of how the Energy efficiency directive is implemented
in the Swedish electricity act. The fourth part is a short survey over the discussions of
tariff design Sweden.
Section 9.2 gives an overview of the various grid tariffs that are offered by Swedish grid
companies. The structure of the tariff provides different incentives for consumers to
modify their consumption patterns. A change in design can have impacts on both the load
profile and the energy use.
Analysing design of grid tariffs is important for two reasons:
New technologies such as interval metering and two-way communication provide
new opportunities for tariff structures.
Different tariff structures give consumers different incentives and can therefore
impact how they use the grid.
New technologies (metering and communication) gives opportunities for a more flexible
use of electricity over the hours during a year. The new technology can therefore make
the customers more price sensitive in the future.7 New metering devices can change the
design for household customers and small firms from traditionally energy based to power
based charging.
9.1 Limited restrictions on grid tariff design in the Electricity Act
Grid tariffs are treated in Chapter 4 of the Electricity Act, which places the following
requirements on grid tariffs:
1 §: Grid tariffs should be objective and non-discriminatory. They should be designed in a
manner compatible with the efficient use of the power grid and an efficient electricity
generation and electricity use.
2 §: Grid tariffs for the transport of electricity shall be structured so, that a paid
connection charge gives the right to use the national electricity grid, with the exception of
interconnectors to neighbouring countries
3 §: On distribution level, grid tariffs for the transport of electricity cannot be structured
taking into account the location of the connection point within the DSO region.
9a §: Grid tariffs for connection to a line or a grid shall be structured so that justifiable
costs incurred by the holder of the concession are covered. On transmission level, the
geographical location of the connection point and contracted power at the connection
point can be taken into account.
7 Most studies of customers price elasticity reports rather low levels of elasticities.
48
5 Chap. 7 a §: The revenue frame should be determined with consideration to in what
extent the network is operated in a way that is consistent with an efficient use of the
network. This evaluation may imply a decrease or increase of the reasonable return on the
capital base.
Summarizing, the Electricity Act only puts two requirements on grid tariffs since the
deregulation decision of parliament in 1995: it should be objective and non-
discriminatory. As long as these two requirements are met, grid companies are free to set
grid tariffs as they see fit. The Inspectorate sets the size of the total allowed revenue, but
does not decide how grid tariffs are structured. Since 2014, the Act also includes the
following provision on grid tariffs: “They should be designed in a manner compatible
with the efficient use of the power grid and an efficient electricity generation and
electricity use”. This change is also supplemented with the formulation in 5 Chap. 7 a §
which gives the regulator a possibility (a measure) to implement incentives in the revenue
frame for a higher capacity utilization. The change in the act regarding tariff design
depends on the requirements in the energy efficiency directive.8
9 Resulting from this
change, the Energy Market Inspectorate has published a provision on how efficient grid
use is determined and how it influences the income caps of the DSOs10
. The provisions
are valid for the regulatory period 2016-2019 and grant higher incomes for DSOs which
reduce network losses and achieve a more even consumption profile.
9.2 Structure of grid tariffs for load in DSO-grids
Energy markets inspectorate (Ei) collects the tariffs from each DSO in Sweden on a
yearly basis. In the beginning of each year a survey sends out to all DSO-grids. They
have to report the structure for a number of customer categories. From the tariff for
apartments (with defined use of electricity of 2000 kWh/year and fuse of 16 A) to an
industry with subscribes power of 20 MW and electricity use of 140 GWh/year.
The structure of tariffs can be downloaded from Ei’s website.11
A description and analysis of tariff structure and the design of tariffs was commissioned
by Ei and done in 2011 by SWECO.12
The report considers the relations of the cost
structure in distribution to tariffs with principles of cost allocation. The cost allocation
has the purpose of fulfilling the goals for objective, cost-reflective tariffs.
In a report to the government in 2012, Ei analysed the design of tariffs in the perspective
of the Electricity Act and the proposal for the Energy Efficiency Directive.13
A typical feature of the structure of tariffs is the variation in design. The structure of cost
is in principle the same, but still there is a rather wide variation in design. In Figure 15,
8 Betänkande 2013/14:NU18 Genomförande av energieffektiviseringsdirektivet.
9 http://www.regeringen.se/sb/d/18263/a/233650
10
http://www.energimarknadsinspektionen.se/Documents/Publikationer/foreskrifter/El/EIFS_2015_6
.pdf 11
http://www.ei.se/sv/Publikationer/Arsrapporter/elnatsforetag-arsrapporter/ 12
http://www.ei.se/sv/Publikationer/Rapporter-och-PM/rapporter-2011/lokalnatstariffer-struktur-
och-utformning/ 13
http://www.ei.se/sv/Publikationer/Rapporter-och-PM/rapporter-2012/
49
the proportion of the fixed part in the tariff for two typical household customers are
shown (with 16 and 25 Ampere and 5 000 and 20 000 kWh electricity). The fixed part
makes up a larger share of the total for the smaller user.
Figure 15 | Fixed part of the grid tariff for user types with 16 and 25 A, 5.000 and 20.000 kWh/year respectively in 2015
Source: Swedish Energi Market Inspectorate
Over the years, there is a slow increasing trend for more fixed part. In 2011, the mean
value of the fixed part was 50 %.
In Table 8, tariffs for years 2011 and 2015 are shown regarding the proportion of fixed
part and the proportion of time-of-use tariffs. As can be seen, there is a variation of the
fixed part from only 1 % for category 20 MW/140 GWh to 73 % for apartments. The
fixed part has grown somewhat since 2011. This trend has been going on slowly since the
deregulation in 1996.
Table 8 | Proportion fixed charge in grid tariff for different customer categories and proportion of Time-of-use tariffs
Level of fuse
Energy use Proportion fixed charge Proportion of Time-of-use
Ampere kWh/year 2011 2015 År 2015
Apartments 16 2 000 72 % 72 % 0 %
16 5 000 67 % 71 % 0 %
20 10 000 64 % 68 % 6 %
Small houses 20 20 000 50 % 53 % 5 %
25 20 000 55 % 59 % 5 %
25 30 000 46 % 49 % 8 %
35 30 000 56 %
35 30 000 55 % 58 % 5 %
50 100 000 38 %
50
63 25 000 72 % 61 % 5 %
80 80 000 54 % 57 % 13 %
>35A 100 100 000 54 % 56 % 11 %
125 125 000 54 % 56 % 17 %
160 350 000 38 % 53 % 19 %
200 240 000 50 %
Power 100 kW 350 MWh
13 % 64 %
1 MW 5 GWh
3 % 64 %
20 MW 140 GWh
1 % 58 %
Source: Swedish Energi Market Inspectorate
Table 9 | DSO grids using power based tariffs for households in 201514
.
DSO grids using power based charging for households customers Excl. VAT and governmental fee (54 SEK)
20 A 20 MWh per year Falbygden Malung Partille Sala-Heby Sollentuna
Fixed charge SEK/year 1.000 2.214 710 1.200 1.200
Power charge SEK/kW and month
Peak-load winter 26,5 72 19,5 79,04 69,6
Off-peak winter 26,5 72 19,5 0 0
Peak-load summer 26,5 22 19,5 32 34,8
Off-peak summer 26,5 22 19,5 0 0
Energy charge öre/kWh Peak-load winter 17,5 0 17,2 0 0
Off-peak winter 17,5 0 17,2 0 0
Peak-load summer 17,5 0 17,2 0 0
Off-peak summer 17,5 0 17,2 0 0
Number of hourly values in monthly charging 1 5 5 3
Source: Swedish Energy Markets Inspectorate
9.3 Discussions on tariff design in industry
With the rolling out of AMR systems for reading the customers’ use of electricity, it is
now possible to use power based charging for household and other small users of
electricity. Therefore there is a discussion in the industry regarding the design of grid
tariffs.
The industry association for grid owners has a working group for tariffs and the group has
written a paper on the issue called “Ny tariffstruktur”. The purpose of the report is to
suggest a structure for cost reflective tariffs. Inputs for the paper are governmental studies
on tariff design with special focus on renewable generation, electrical vehicles and a more
efficient energy15
, as well as the Energy Efficiency Directive from EU. The discussion
regarding a common Nordic retail market (with proposal of a supplier centric model) is
also an input to the work.
14
SOU 2005:51 “Bilen, biffen, bostaden. Hållbara laster – smartare konsumenter”, 2005
SOU 2008:110, ”Vägen till ett energieffektivare Sverige”, 2008.
15 Ei 2012:14, ”Tariffutformning - behövs mer detaljerade krav”, 2012.
51
The focus has been to discuss how a power based tariff can be designed. In short, the
group presents the following suggestion for a future structure:
A fixed part and two variable parts with charging both on use of electricity and the
maximum load during each month. The billing is on monthly basis on actual use and load.
The fixed part will cover customer specific costs such as metering and billing and
customer service. The power part is metered as the highest monthly hourly load (kWh/h) -
that is the highest hourly mean value of load. The charge will cover capacity costs in the
local grid but also the local grid’s costs for upstream grid (sub-TSO). The third part is a
distribution charge16
based on the monthly use of electricity (kWh). This part will cover
the costs for grid losses both in the own grid and upstream grid. Some capacity cost for
central parts of the local grid will also be incorporated in the distribution charge. The
reason for that is that some customer groups have collective evening out of the loads.
This is the case for small users with low utilization time like apartments.
The reason for the proposal is threefold:
All customer groups shall participate in paying for the ground costs (cost which is
not depending on power or energy)
All customer shall have the same definition of power value
To use the mean value of several hourly values for charging is not recommended
(which is the case for some grids in Sweden)
The group also suggest that the power charge shall be the same during all hours during
the year – that is a uniform tariff and no time-of-use tariff. If a time-of-use charging will
be used, it should be based on energy use (kWh).
Tariffs for apartments with good power simultaneity has in some cases a completely fixed
part or a fixed and energy based part as the only customer group. All other customer
groups have power based charging. The group is of the opinion that in the future, even the
apartment customers shall have a power based charging. The reason for that is greater
variation in the load profile between customers in apartments.
In a report from a tariff group, eight local DSO’a in the north of Sweden give a proposal
for a new tariff.17
They call it a “market monitored tariff” meaning that the grid tariff
depends on the demand for capacity (kW). This tariff is adjusted each year with changes
in demand for capacity. The charge for power will be different for each month depending
on the load (capacity utilization), so there will be twelve different prices for power
(SEK/kW). The load profile is calculated from the collective load – that is – all customers
have a responsibility for the collective load. The tariff can therefore be characterized as a
seasonal tariff because the load is much higher in the winter due to electricity heating. For
some of the grids, there are customers visiting the mountains for skiing, further increasing
the demand for capacity during these times.
The calculation of the tariff is based on the maximal load each month for the last two
years. The load profile is used for the allocation of costs.
16
Överföringsavgift (öre/kWh). 17
Elinorr – tariffutredning – delprojekt 1, 2014.
52
There are several objectives for tariff setting:
Cost reflectivity
Simplicity
Giving incentives for efficient use of capacity. The customers shall have incentives
to even out their load over the day.
Stable revenues
Less administration
The proposal gives a different price/kW load each month, but not a different price over
the day or weekends. They propose a seasonal time differentiations, but not within the
month. An explanation for that position is that these local grids have a special high load
during the winter months and a time-of-use tariff over the day will probably give high
loads during the night – that is – shifting the loads from day do night, but still high loads.
9.4 Conclusion
The Energy Efficiency Directive has been implemented in the Swedish Electricity Act.
The supplement adds a wording that the tariff must give incentives for efficient use of the
grids, but also incentives for efficient production and use of electricity.
The grid industry has the opinion that a change from fuse tariffs to power based tariffs is
good. A possible exemption can be apartments, but several DSOs think that even
apartments ought to have power based tariffs, possibly with another pricing structure
between fixed and power based charging compared to other customer groups.
The industry is more hesitant to implement time-of-use tariffs over the day and weekends.
If there shall be a time differentiation, they think that the differentiation shall be on the
energy part (öre/kWh). Some grids think that a change to a power based tariff can be done
in two steps. First a uniform power based tariff (no differentiation) and then after some
years a second step with time differentiation. Today, there are only three local grids using
a complete time-of-use tariff for household customers (like electricity heated houses).
Two other local grids have power based charging with uniform price over the year.
NordREG
c/o Danish Energy Regulatory Authority
Valby 2500
Carl Jacobsens Vej 35
Denmark
Telephone: + 45 41 71 54 00
Internet: www.nordicenergyregulators.org